Project Description Functional MRI (fMRI) based on the blood oxygenation level dependent (BOLD) contrast has become a powerful neuroimaging modality and has gained a prominent position in neuroscience for imaging brain activation at working state and functional connectivity at rest. However, most of fMRI research focus on functional mapping of brain activity at the system level with macroscopic scale. Recently, high-resolution fMRI at ultrahigh field has shown the feasibility of mapping the functional activity of elementary computational units from ocular dominance to orientation column. Such unprecedented neuroimaging ability opens up exciting opportunities for studying brain function, connectivity and circuitry at the mesoscopic scale. Nevertheless, the neural computational processes are distributed across six cortical laminae spanning from the pial surface to the white matter, and engage feed-forward, feed-backward and local connections that are segregated according to the cortical depth. Ability to map such laminar and columnar dependent functionality and connectivity across large networks is extremely challenging and has not been achieved to date. Moreover, the BOLD signal only reflects the secondary effect of neuronal activity, the transformation between the BOLD measure and the underlying neural activity becomes complicated at varied spatial scale, and the neuro-BOLD correlation at the laminar/columnar level has not been studied due to a variety of technical hurdles. Another highly relevant unanswered question in fMRI is how does neuronal inhibition change the neural dynamics and networks, and the fMRI BOLD signal. Owing to the high complexity of normal brain activities unavoidably involving both excitatory and inhibition processes, it is a daunting challenge to selectively study the neural correlate of BOLD to inhibitory neuromodulation. To address these questions and challenges, this proposal aims to push the technology envelope beyond the current level by developing innovative multimodal fMRI approaches capable of simultaneous neural stimulation, recording and fMRI acquisition with functional mapping specificity and resolution down to the mesoscopic scale. The cutting-edge technology and developed tools will allow us to investigate brain function and connectivity at cellular columnar and laminar levels?two most fundamental neural computational units for micro-circuits essential for brain function, and still cover large networks through thalamo-cortical and cortico-cortical connections in the cat brain. For the first time, the research will provide new knowledge about the neural dynamics in space and time, and neural correlates of fMRI BOLD signal in response to excitatory or inhibitory neuromodulation at laminar/columnar levels. Such knowledge is impossible to gain from the human brain research, but should lead to transformative breakthroughs in understanding the structure-function relationship of defined computational units, dynamic functions and networks of the human brain; and provide new insights into electrophysiology basis and mapping specificity of fMRI at the laminar and columnar levels.